I am looking for the name for the logical fallacy that compares the best samples from one group with the worst samples from the second group and concludes that the first group is better than the second. Example (maybe a little vague): In a gun-related debate a picture is presented with two doctors who were recently accused of murder and a person with a gun (probably a marine) and the question is asked: "who is the real murderer?" The conclusion that one is supposed to draw is that based on this comparison of "the worst" of the doctors vs. "the best" of the gun users is that the whole group of gun-users is "good".
Informal fallacies have fuzzy boundaries. This seems to me to be a variation on "cherry picking," which is basing a general conclusion on a few carefully selected data points (which may not reflect the larger trends).